Friday, November 30, 2018

Hoping to defend the iPhone against Qualcomm, chip designer waited for almost 12 hours on hard wooden seat at Munich courthouse

Litigation is a hardass business, as I'm sure none of this blog's informed readers would deny. At the Munich I Regional Court ("Landgericht München I"), one of Europe's patent powerhouses, that's also literally true. Should the court ever adopt the German custom of naming benches after real persons (you can see the little name plates in many parks), I propose the following seating area right outside of courtroom 501 be named after a certain Mr. Mike Kay since he spent almost 12 hours waiting there on November 8, unfortunately in vain, for an opportunity to testify in a Qualcomm v. Apple patent infringement trial:

Mike Kay is a Qorvo employee and designed the envelope tracker chip used in some iPhones models that Qualcomm alleges infringe on its European patent EP2724461 ("low-voltage power-efficient envelope tracker"). The court must be exonerated for a couple of reasons. First, because Mr. Kay was so eager to speak out on the inner workings of his chip, he voluntarily stayed as close to the courtroom as possible, though at the other end of the hallway there would have been a slightly more hospitable seating area:

What's more important is that the defending alliance (Apple-Intel-Qorvo) brought him along to the November 8 trial without the court having put him on the official witness list. There was no formal discovery order in place. German courts can, in their discretion, depose a witness who happens to be around voluntarily. Presiding Judge Dr. Matthias Zigann didn't exercise his discretion to that effect, but he's not responsible for rewarding perseverance. He could, however, still do so in a decision currently scheduled to be announced on December 20, provided that the three-judge panels he presides over will conclude that it's necessary in order to arrive at a well-founded, factually correct judgment. One of the things that may give the judges pause is the fact that an Administrative Law Judge (ALJ) at the United States International Trade Commission (USITC, or just ITC), with the benefit of Mike Kay's testimony and access to the schematics of the accused circuitry, concluded that Qorvo's chipset architecture does not infringe a claim of a U.S. patent from the same Qualcomm patent family.

I spend most of my time on things unrelated to this blog and the topics it covers, and that was also the case on the trial date. However, when the story of "Mike's ordeal" made the rounds in the IP community, I tried to find out more. A spokeswoman for the court pointed out that Mr. Kay was not on the court's witness list for that particular trial date. Apple declined to comment; Qualcomm's great PR agency (Burson Cohn & Wolfe) was quick to respond, but didn't elaborate beyond correcting a factual error (regarding a technical aspect of the case) in a news agency article. So I reached out to Intel, who reacted the way Apple did, and Qorvo. I very much appreciate the competence and the patience with which Qorvo's Chief IP Counsel Michael Baker, who had attended the trial, answered my questions. I didn't have the opportunity to talk to Mike Kay, but got all the information I needed.

Mr. Baker stressed that Qorvo actually has a good relationship with Qualcomm in some other areas, but they do disagree with Qualcomm's infringement assertions in this case. And he convincingly explained that Qorvo has the greatest respect for both the Munich judges and the renowned, court-appointed chipset expert, Professor Stefan van Waasen. However, Qorvo believes the court and Dr. van Waasen should have all of the information--the chipset schematics as well as Mike Kay's testimony--in front of them when they opine and rule. Mr. Baker told me Qorvo would be more than happy to present the schematics to the judges, the expert, and parties' counsel in order to prove that various claim limitations, with particular emphasis on an "offset" (number 328 on Figure 5 of the patent specification), are not practiced by Qorvo's envelope tracker chip. They say their product achieves the same power-saving effect (which is relevant when battery power goes down) by different means, but in this "hypercompetitive" market (as Mr. Baker describes it) they can't possibly let Qualcomm's or anyone else's engineers take a look at their schematics.

According to a German news agency report that mischaracterizes the technical aspects of the case, Dr. van Waasen basically said he can't see how else (if not by infringing the patent-in-suit) the accused iPhone models achieve the same effect. But that opinion was based on a tear-down report provided by Qualcomm, and as I've been told, such tear-down reports are inherently less than 100% reliable, much less complete.

Most of this blog's readers are based in the U.S. and may wonder: why can't they just solve this through an "AEO" (Attorneys' Eyes Only) designation of the material in question and exclude the public from Mike Kay's testimony?

But in a similar way Germans might ask: why the hell do they need juries in the U.S. and multiple weeks of testimony to rule on patent infringement questions that German courts quite often adjudge after a two-hour trial without any witnesses or discovery? There simply are differences between the jurisdictions.

German courts can also depose witnesses behind closed doors. However, there's no German equivalent to a protective order.

Actually, everything was in place and on track to get the best of both worlds: the efficiency of German proceedings and the flexibility regarding a protective order provided by U.S. law. But then Qualcomm made an about-face that I really don't understand. I've downloaded and studied various documents from a U.S. discovery proceeding that Qualcomm started in March, telling the U.S. court it needed to conduct discovery of Qorvo's chipset architecture under 28 U.S.C. § 1782. In the summer, the United States District Court for the Middle District of North Carolina (where Qorvo is headquartered) approved a protective order that Qualcomm and Qorvo had negotiated (this post continues below the document):

Stipulation and Protective ... by on Scribd

As far as I understand, while Qualcomm's U.S. counsel from the Quinn Emanuel firm agreed to the protective order, Quinn Emanuel's German lawyers elected not to do so, and thus claimed they couldn't present Qorvo's chipset schematics at the November 8 trial.

This is evidentiary minimalism on QE's part. It's in stark contrast to what the firm's founder, John Quinn, told Judge Lucy Koh of the United States District Court for the Northern District of California at a 2014 Apple v. Samsung trial. "What's the point in having a trial [without all the evidence on the table]?" I agreed with Mr. Quinn then. I agree with Qorvo, between the lines, saying the same thing now.

I don't have, and for lack of access to the evidence, won't ever have an opinion on whether Qorvo's chip practices all limitations of the patent claim-in-suit. But the outcome at the ITC does suggest the schematics and Mike Kay's testimony are relevant, and there's a pattern here where Qorvo goes to extreme lengths, as long as the confidentiality of its chip design be protected, including that they offered to set up a complete test environment in Germany for the expert, while Qualcomm and Quinn Emanuel, for whatever reason or no reason, prefer the expert to opine on the basis of a reverse-engineering effort commissioned by Qualcomm.

Also, I've been party to a couple of non-patent cases in Germany where courts appointed experts, and the ordinary procedure here is that an expert furnishes an opening report, enabling the parties to ask questions that give rise to one or (often) more supplemental reports before the expert finally testifies in court. Here, Judge Dr. Zigann made it clear in February that the appointment of an expert only for the November trial was going to be an experiment. I can understand that he, actually a leading author on patent procedures, sought to streamline the process. With multiple rounds of questions to be put before the expert, it could have taken another 6-12 months to resolve this case. But even the 12-hour court session on November 8 (from 9 AM to 9 PM local time) apparently failed to shed light on some key technical aspects, for lack of access to the material Qualcomm's counsel had actually obtained through a U.S. discovery proceeding.

There's a certain indication that the German court wasn't necessarily satisfied with the state of affairs at the end of the November 8 trial as the official minutes of the meeting (based on what I learned) explicitly mention that the court would entertain a motion to reopen proceedings. Since no leave to file such a motion is required under German procedural law, that may mean that a reopening of the record may happen contrary to a non-final holding at the trial that the defendants presented certain evidence out of time. As far as I understand things based on what Qorvo told me, the timing problem was due to Qualcomm's about-face as they had all originally relied upon the chipset schematics (and Mike Kay's testimony) forming the basis of the decision in Germany, just like at the ITC.

In all the patent infringement cases I watched, a reopening of proceedings (which basically means that another trial is scheduled) was needed at least once. The Mannheim Regional Court reopened a Motorola Mobility v. Microsoft case over a push notification patent that ultimately turned out invalid anyway. Should the Munich Qualcomm v. Apple case over Qorvo's envelope tracker continue, I have no doubt the court won't let Mike Kay wait outside for so many hours. And, in the hypothetical, I'd recommend to Mr. Kay to sit on the softer chairs down the aisle, with a nicer view.

Share with other professionals via LinkedIn:

Tuesday, November 20, 2018

Call to action: Munich should remain a leading patent litigation venue in Europe

Roughly two thirds of all European patent infringement cases are brought in Germany. Unlike in the U.S., where patent cases can be filed with any district court in the country, only a limited number of German courts have in rem jurisdiction over such cases, and only three of them really matter: Düsseldorf (this venue gets most cases, but not in the smartphone industry), Mannheim (the primary smartphone venue, where some judges almost deserve an honorary doctorate in radio frequency electronics), and Munich, where I grew up though I'm westbound by now.

Munich has two regional courts. Munich I has in personam jurisdiction over cases involving actions or persons within the city border, and in rem jurisdiction over patent cases; Munich II serves the outskirts (and doesn't try patent cases).

Munich I earned its place among the top three German patent venues--and, I would say, among the top five in Europe--for a combination of reasons:

  • It used to be a rocket docket, about as speedy as Mannheim and clearly faster than Düsseldorf.

  • It has its own--I'll use a U.S. term--patent local rules. Unlike Mannheim and Düsseldorf, where everything substantive is discussed in a single trial, Munich has broken the trial up into two parts, an "early first hearing" ("frühe erste Anhörung") and a "main hearing" ("Hauptverhandlung"). The first hearing usually takes place a few months after the complaint was served, and is somewhat similar to a U.S. Markman hearing, though they usually don't write down a claim construction the way U.S. courts do and, beyond the Markman scope, start to discuss infringement questions. The early first hearing gives parties a chance to fine-tune their argument with a view to the second hearing, which is the actual trial.

  • Many patent-focused firms have a strong presence in Munich (for an example, Quinn Emanuel's German center of gravity appears to be shifting from Mannheim to Munich) because of the strength of the regional economy (including many subsidiaries of major U.S. tech companies) and all those proceedings taking place before the European Patent Office (oppositions to recently-granted patents) and the Federal Patent Court of Germany ("Bundespatentgericht"; nullity proceedings against German patents or, typically, German parts of European patents).

  • While this is only a minor factor, Munich is also far more of a tourist attraction than Düsseldorf, let alone Mannheim, which is almost creepy and has an increasing problem with violent crime.

But there's a serious problem--a lack of political support for this patent venue.

While the government of the state of North Rhine-Westphalia, regardless of whether the state is run by conservatives in name only (as it is now) or by the self-declared political left (as it was before), recognizes patent infringement litigation as a regional economic development factor, the party that has been in government in the state of Bavaria for half a century (CSU) appears to be pretty clueless, which is irreconcilable with its "Laptop und Lederhose" (laptop and Oktoberfest-style leather pants) slogan. Instead of strengthening the "civil law chambers" ("Zivilkammern") that hear patent cases in Munich, the court's former chief judge even reduced staff size by one judge, which sounds like a minor difference but has huge practical implications whenever one of the three judges (and they need three to form a panel that can hear and adjudicate a case) is on vacation or ill.

That's why I'm asking those of you who have a professional interest in Munich remaining a major patent litigation venue, also with a view to the future Unified Patent Court (UPC), to help those provincial folks figure out the problem and, more positively speaking, the potential.

A new Bavarian state government has just been formed, and a new state AG (again I used a U.S. term; in German, he's called "Justizminister", or "minister of justice"), who is ultimately in charge of providing the Munich I court (unlike a U.S. district court, a German regional court is a state--not federal--court) with sufficient resources, has just been appointed. Georg Eisenreich was previously in charge of (among other areas of responsibility) "digitization," a fact that may make him much more receptive to patent-related issues than the average state-level politician.

You can find his ministry's contact data on this web page. If you care about Munich as a patent litigation venue, please write to him and explain that you've learned about a recent staff cut affecting the "7. Zivilkammer" of the Landgericht München I, and that you would like to express your concerns about how this decision, beyond its practical implications, sends out a signal that threatens to reduce the relevance of Munich as a patent litigation venue. You may wish to highlight the following points:

  • It's important for Bavaria as a state that prides itself on its strong technology sector to have a strong patent infringement court.

  • Patent infringement cases tend to generate very high court fees for the state, so strengthening--not weakening--the venue will pay for itself, and more.

  • The state government should strive to ensure that Munich remains a significant venue if and whenever the Unified Patent Court starts.

  • Many patent litigants travel to German patent litigation venues from other countries (very often even from other continents). The government of the state of North Rhine-Westphalia appears to be aware of the economic benefits of such business travel, and so should its Bavarian counterpart.

  • Patent prosecution and litigation firms have created, and will continue to create, many high-paid jobs in the Munich area, some of which depend on the ability to enforce patents in their backyard. Also, many inventors may seek patent attorneys in cities they read about in the context of high-profile patent litigation involving U.S. and Asian technology giants.

  • If the state government fails to act, other specialized judges, whom it is hard to replace, may follow the example of former presiding judge Dr. Peter Guntz, who was hired away by the European Patent Office, where his after-tax income is presumably a lot higher.

None of this is meant to criticize the work performed by the court's patent judges. It's all about what the state government should do in order to let those specialized judges do their work as efficiently as their peers in other major German patent litigation venues.

Share with other professionals via LinkedIn:

Sunday, November 11, 2018

More than a dozen public interest statements filed in investigation of Qualcomm's first ITC complaint against Apple

In late September, the ITC's meanwhile-retired Administrative Law Judge (ALJ) Thomas B. Pender recommended that the U.S. trade agency refrain from banning Intel-powered iPhones he deemed to infringe a Qualcomm patent, given Qualcomm's overtly anticompetitive litigation tactics of targeting only Intel-powered iPhones. In late October, a heavily-redacted version of his findings became available. On Halloween, the parties filed their public interest statements.

The parties, government agencies, and the general public have multiple opportunities to submit public interest statements to the ITC. I've created a three-page diagram that shows at which procedural milestones the ITC requests and/or invites such statements.

On Thursday (November 8), the latest round of statements by the general public was due. A total of thirteen statements were filed that day. Here are some observations:

Intel's statement is particularly relevant. Its testimony on what would happen if Qualcomm could successfully exclude Intel-powered iPhones from the U.S. market is at the heart of ALJ Pender's factual findings relating ot the public interest.

Intel's public interest statement reinforces its testimony (this post continues below the document):

18-11-08 Intel Public Inter... by on Scribd

The Computer & Communications Industry Association (CCIA) is an industry body whose members include many major hardware and software companies (including Intel). CCIA supports ALJ Pender's recommendation against a U.S. import ban of Intel-powered iPhones.

ACT | The App Association speaks for many app developers and receives funding from some large corporate sponsors including Intel. ACT's statement contains the following passage:

"And the fact that Qualcomm is seeking an exclusion order only for Apple's iPhones that contain Intel modem chips, and not all iPhones, strongly suggests that damages available in the district court for any infringement of its patents should be adequate compensation."

It's true that an Apple competitor like Samsung or Huawei wouldn't discriminate between Qualcomm-powered and Intel-powered iPhones (except if some license agreements required differentiation). But Qualcomm just wants 100% of Apple's baseband chipset business.

The joint statement of the Electronic Frontier Foundation (EFF), Engine Advocacy, the R Street Institute, and the (conservative/libertarian) Lincoln Network connects the dots between Qualcomm's behavior in its dispute with Apple in the Southern District of California (where it's willing to give up on a host of patents just to avoid a ruling on the question of patent exhaustion) and its insistence on an ITC import ban over a single patent:

"Second, there is strong reason to believe that Qualcomm is not even interested in compensation for any patent infringement. In a related district court case, Apple sought declaratory judgment that it did not infringe several Qualcomm patents. Rather than countersuing for infringement, Qualcomm voluntarily gave Apple a covenant not to sue for infringement. Certainly the patents at issue before the Commission are different, but Qualcomm's willingness to forego any remedy on a large portion of the company's patent portfolio strongly suggests that recompense for its patent rights is not Qualcomm's motivation."

Public Knowledge and the Open Markets Institute focus on the discriminatory nature of Qualcomm's decision to target only Intel-powered iPhones:

"Qualcomm's anticompetitive intent is obvious. [...] Because many of the disputed patents are common to all iPhones and do not concern baseband technology at all, it’s clear that Qualcomm is choosing to selectively enforce its patents in a way designed to maintain its monopoly position in the baseband market."

But this time around, unlike in the context of the FTC's successful summary judgment motion, Qualcomm has also received broadbased support from stakeholders.

France Brevets (a governmental patent troll) is obviously interested in maximizing patent holders' leverage, as are IP Europe (Qualcomm is a member of its "Global Innovation Taskforce") and the Innovation Alliance (co-founded by Qualcomm), which pretty much represent the interests of the Qualcomms, Nokias, and Ericssons of the world, and Inventors Digest, a magazine for patentees.

Then there are four conservative groups supporting Qualcomm. In three of those cases it's hard to reconcile those organizations' names with what they're advocating here:

  • Americans for Limited Government ("if they were asking for a subsidy or hand up, we would likely oppose it") wants a government agency, the ITC, to instruct another government agency (Customs & Border Protection) to give Qualcomm leverage over its competitor

  • Frontiers of Freedom says "[i]t is outrageous that a judge can conclude that a company has violated the patent rights of another company but then impose no punishment or remedy whatsoever," when it's just due to the statute that the ITC doesn't have more flexibility regarding remedies. In any event, import bans are not a matter of freedom.

  • Conservatives for Property Rights submitted its IPWatchdog article and complains about "the highly irregular outcome in this matter" without attributing it to Qualcomm's highly unusual litigation tactic of specifically targeting Intel-powered iPhones. But more fundamentally, what about the property rights of Apple and Intel--two companies that definitely have invested and continue to invest a lot in innovation?

What disappointed me more than anything else is that an organization I actually like a lot--the American Conservative Union, famous for organizing CPAC--ignores the unusual parameters (due to Qualcomm's litigation tactics) of this ITC investigation and blows things completely out of proportion:

"A recent ruling at the International Trade Commission (ITC) just put all American companies on notice: 'We are all socialists today.'"

That term reminded me of a story that is exactly 13 years old. A then-SAP executive described open-source software as "IP socialism," and I disagreed then as I do now.

First, neither ALJ Pender nor Apple said Qualcomm's innovations should be socialized. The question here is just about the appropriate remedy. It's not about whether or not Qualcomm would get paid. If it prevailed in district court, it absolutely positively would be entitled to damages.

Second, it's perfectly conservative to prevent companies from monopolizing markets. Two great Republicans from Ohio were key to the creation of antitrust law: Senator Joseph Sherman sponsored the namesake law, and President Benjamin Harrison signed it into law. In its 1993 Spectrum Sports v. McQuillan decision, the Supreme Court explained what the Sherman Act always was about: "The law directs itself not against conduct which is competitive, even severely so, but against conduct which unfairly tends to destroy competition itself."

The public-interest issue in ITC investigation no. 337-TA-1065 is about preventing Qualcomm from leveraging patents to destroy competition itself. Conservatism is all about fair competition, not about its destruction.

Share with other professionals via LinkedIn:

Saturday, November 10, 2018

Judge Koh's high-profile summary judgment order doesn't bode well for Qualcomm's efforts to elude patent exhaustion finding

This is a follow-up (as I promised) to Judge Lucy Koh's summary judgment order according to which Qualcomm must meet its self-imposed obligation to license its cellular standard-essential patents (SEPs) to rival chipset makers. On Tuesday I mostly wanted to publish the news quickly, and I focused on the commercial consequences.

The legal standard applied by Judge Koh was stated as follows in the Ninth Circuit's 2006 opinion in Miller v. Glenn Miller Prods., Inc.: summary judgment is warranted "[i]f, after considering the language of the contract and any admissible extrinsic evidence, the meaning of the contract is unambiguous." (emphasis added)

Qualcomm unsuccessfully argued that an alleged need for two U.S. standard-setting organizations (ATIS and TIA) to be consistent with policies established by other organizations supported its position that there was no obligation to extend a SEP license to rival chipset makers, and that baseband chips don't actually implement the standards in question.

Not only did Judge Koh find that the FRAND licensing pledges Qualcomm had entered into were unambiguous, but she also saw that a right to withhold SEP licenses from competitors would allow someone to monopolize a market, which is precisely what all those FRAND declarations are meant to avoid. And she drew the inevitable conclusions from how Qualcomm and its executives had previously characterized their chips as "the heart of a cellphone," or testified that key cellular technologies were "implemented" in modem chips. The following two passages are particularly important:

"Despite having SEP licenses for its own modem chips, Qualcomm argues that its FRAND obligations for SEPs extend only to device suppliers and not modem chip suppliers because only device suppliers 'practice' or 'implement' standards. However, that distinction not only violates the non-discrimination obligation, but also makes little sense. As Qualcomm's founder conceded and Qualcomm's own documents demonstrate, modem chips may be 'compliant' with cellular standards."

"Moreover, undisputed evidence in Qualcomm's own documents demonstrates that a modem chip is a core component of the cellular handset, which only underscores how a SEP license to supply modem chips is for the purpose of practicing or implementing cellular standards and why Qualcomm cannot discriminate against modem chip suppliers. In an amicus brief filed in the Federal Circuit, Qualcomm characterized its own modem chips as 'the heart of a cellphone.' [...] Qualcomm's founder testified in a deposition that key cellular technologies were 'implemented' in modem chips. [...] In Qualcomm's own Annual Report, Qualcomm stated that Qualcomm is a 'leading developer and supplier' of circuits, including modem chips, 'based on' the CDMA family of cellular standards. [...] Qualcomm also represents that Qualcomm's modem chips 'perform the core modem functionality in wireless devices.'"

Those statements will presumably also bear significant weight with the appeals court.

From Qualcomm's point of view, Judge Koh's conclusion that baseband chips are indeed the heart of a cellphone (and implemented cellular SEPs) is unhelpful beyond the FTC v. Qualcomm case and the desire of companies like Intel to secure a FRAND license to Qualcomm's SEPs:

  • Qualcomm is trying hard to avoid a ruling on whether its SEPs are exhausted by chipset sales, but if and when a court looks into this question, Qualcomm will argue that baseband chips don't implement standards all by themselves--just like Qualcomm did in its opposition to the FTC's summary judgment motion.

  • Patent exhaustion and standard-essential patents are not at issue in the investigations of Qualcomm's two ITC complaints against Apple, but in connection with the ITC's statutory public interest factors (see this recent post on the related procedures) Qualcomm also argues that the focus should be on entire mobile devices, not baseband chips. Qualcomm's ITC complaints target only iPhones without a Qualcomm baseband chip, which in practical terms means Intel-powered iPhones.

The order Judge Koh handed down on Tuesday is likely to be cited a lot going forward.

Finally, there are three recent media reports related to the Qualcomm antitrust and patent matters that I'd like to draw your attention to:

Share with other professionals via LinkedIn:

Tuesday, November 6, 2018

Qualcomm must license patents to rival chipset makers: Judge Koh grants FTC's partial summary judgment motion


Judge Lucy H. Koh of the United States District Court for the Northern District of California has just granted the Federal Trade Commission's motion for partial summary judgment against Qualcomm regarding the latter's self-imposed obligation to license, on FRAND (fair, reasonable and non-discriminatory) terms, its cellular standard-essential patents (SEPs) to rival chipset makers such as Intel (this post continues below the document):

18-11-06 Order Granting Par... by on Scribd

The motion was strongly supported by the rest of the industry, apart from Nokia and Ericsson (companies that actually used to take the same position against Qualcomm many years back).

As the FTC clarified in response to Nokia's filing, the motion was specific to FRAND licensing promises Qualcomm made to two U.S. standard-development organizations--the Telecommunications Industry Association (TIA) and the Alliance for Telecommunications Industry Solutions (ATIS)--as opposed to arguing that all FRAND licensing promises must be construed as benefiting rival chipset makers. In this FTC v. Qualcomm antitrust case pending in Northern California, the FTC also took the position that Qualcomm had the same obligation under the ETSI (European Telecommunications Standards Institute) FRAND declaration, but the FTC sought (successfully, as we know now) to simplify and streamline the case by obviating the need for interpreting a document under French law when the relevant obligation, as Judge Koh has agreed, already exists under FRAND declarations Qualcomm made under U.S. law.

This is the outcome I had predicted. I've said all along that the FTC had a very strong case, with this particular motion for partial summary judgment having represented a sweet spot in terms of focusing on an issue that the court can resolve ahead of trial while tackling one of the most problematic aspects of Qualcomm's (and, to be fair, not only Qualcomm's) conduct.

Nothing made the importance of this motion clearer than a procedural motion brought by the FTC and Qualcomm asking the court to go forward with all other aspects of the case except this one, which the parties didn't want to have adjudicated before November 14. Fortunately, Judge Koh denied that administrative motion right away. In a subsequent post on FTC officials who are against this case I wrote that the likelihood of the motion being granted had increased.

Judge Koh has now made a decision that will presumably result in some other companies, such as Intel, telling Qualcomm again that they want a FRAND license to its cellular SEPs. I guess those renewed requests will happen rather shortly.

What's hard to analyze from the outside is the impact of this on settlement talks between the FTC and Qualcomm. Just a few days ago the Capitol Forum (a subscription service) broke the news that FTC chairman Joseph J. Simons is recused from the matter. This successful motion is, in its own right, a major accomplishment by the FTC's litigators. Qualcomm will definitely want to avoid the January antitrust trial, but the FTC is on the winning track.

Today's order also strengthens the position of consumer plaintiffs in their class action against Qualcomm (in which they're seeking $5 billion in damages, which would have to be paid out to up to 250 million consumers). The fact of the matter is that Qualcomm consistently refused to extend a SEP license to rival chipset makers, but the market would definitely have been more competitive if Qualcomm had complied with the obligation it actually had (according to Judge Koh).

I wanted to publish the decision immediately, and I'll read it carefully tomorrow and may do a follow-up post then.

Share with other professionals via LinkedIn:

Sunday, November 4, 2018

How and when the ITC analyzes its statutory public interest factors: three-page diagram

The roughly 100 pages on which meanwhile-retired Administrative Law Judge (ALJ) Thomas B. Pender discussed the statutory public interest factors in the investigation of Qualcomm's first ITC complaint against Apple warrant further commentary. There are antitrust cases pending against Qualcomm around the globe, and Apple raised an affirmative antitrust defense against Qualcomm's German patent assertions, but ALJ Pender's public-interest analysis is the first detailed judicial holding--based on extensive briefing and a multi-day hearing.

That's why I will soon dedicate a blog post to the subject of inhowfar ALJ Pender's public-interest analysis overlaps with key antitrust questions at issue in various other proceedings. But before we get there, I wanted to provide some procedural background.

It's important to understand that the United States International Trade Commission, a trade agency with quasi-judicial powers, clearly delineates the question of a "violation" (typically, the infringement of a patent not shown to be invalid; I sometimes refer to this as the "merits" part) from "remedies and bonding" (which I'll briefly refer to as "remedies"). This is one major difference from the public interest eBay factor, which is considered together with the other three factors.

The ALJ in charge of an investigation finishes his job by issuing an Initial Determination (ID) on the alleged violation and a Recommended Determination (RD) on the remedy sought. The Commission (the six-member decision-making body at the top of the organization) makes the final (apart from appeals or a Presidential veto) decision on both, but there is a technical difference: the ID becomes the final decision if the Commission declines to review it (which parteis can request); the RD is just a recommendation, and the Commission makes the first actual decision on remedies, which is why parties cannot seek a review of the RD (though they can, of course, voice any disagreement).

And then there's also the ITC staff (formerly called Office of Unfair Import Investigations, abbreviated as OUII). The staff participates in many investigations as a third party (though it sometimes elects not to do so, or limits its participation to specific issues). The staff's involvement is also meant to protect the public interest, but not in the sense of playing a "devil's advocate" and arguing against an import ban. In connection with the staff, "public interest" basically means to file third-party pleadings in order to increase the likelihood of a correct decision (ALJs adopt many, but by far not all, staff recommendations). As far as the statutory public interest factors that may counsel against an import ban are concerned, the ITC staff is normally (with the investigation of Qualcomm's second ITC complaint against Apple representing a remarkable exception) rather patentee-friendly. For the purposes of this post and the diagram contained in it, "public interest" refers to the statutory public interest factors, not the public interest in correct decisions.

The statutory public interest factors ("the effect of [a hypothetical import ban] upon the public health and welfare, competitive conditions in the United States economy, the production of like or directly competitive articles in the United States, and United States consumers") have gained importance over the years. It's an irony of fate that a case where Qualcomm was the respondent (and Broadcom the complainant) played a key role in the evolution of the public interest debate at the ITC. The purpose of those public interest factors is not to condone patent infringement: patentees can always (and do most of the time) file a companion lawsuit in district court and seek an injunction as well as monetary damages there (though it will take longer, and injunctive relief will be far harder to obtain).

When the public interest is at issue, government agencies and the private stakeholders must also be able to chime in--and the parties are obviously listened to as part of this. Before an ITC exclusion order (= U.S. import ban) comes down and actually takes effect, there are various junctures at which the statutory public interest factors may be considered based on input from the parties as well as, potentially, government agencies and the general public. There's an explanation on the ITC website, but I wanted to add the Presidential review period to it and created a diagram (three-page PDF; this post continues below the document):

18-11-04 ITC Public Interes... by on Scribd

Now let's go through the steps shown in the diagram and look at them in the specific context of the ITC investigations of Qualcomm's two complaints against Apple (and, by extension, Intel, as Qualcomm decided to target only Intel-powered iPhones, even though one of the patents is about graphics and, therefore, technically not related to Intel's baseband chips by any stretch of the imagination).

Complainants are nowadays required to file a public interest statement along with each complaint. When a complaint is received, the ITC always solicits public interest statements from the general public. Responds can and often do file their statement at this stage, where it's still voluntary.

What's relatively new is that the Commission decides on whether to delegate fact-finding related to the statutory public interest factors to the ALJ in charge of the investigation. The ALJ will hold an evidentiary hearing either way, but only if the Commission directs him to gather evidence relating to the public interest will he also hear testimony related to those factors (and unrelated to the merits part). And if the ALJ is tasked with this, then the respondent (the company against which someone is seeking an import ban) is required (not merely invited) to file a public interest statement before the investigation is in full swing.

In the bottom third of page 1 of the PDF you find the rectangular box labeled "Public interest briefing, discovery, evidentiary hearing (before ALJ)" and, to the right of it, a reference to the ITC staff presenting its public-interest findings. That is exactly where the investigation of Qualcomm's second complaint against Apple stood when I reported in mid-September that the ITC staff, in the public part of the evidentiary hearing, recommended that no import ban be ordered should a violation (contrary to the staff's analysis, according to which there is none) be identified. However, in the investigation of Qualcomm's first ITC complaint against Apple, the ITC staff sided with Qualcomm. For lack of access to all the relevant documents I don't know why, but the record may differ, just like Presiding Judge Dr. Holger Kircher of the Mannheim Regional Court acknowledged substantial progress with Apple's antitrust defense at an early-October trial over what Apple had presented in a case involving another patent that was tried two weeks earlier.

The investigation of Qualcomm's first ITC complaint against Apple started several months earlier than the second one, and the RD (which will probably take until January inthe other investigation) already came down in late September. That procedural milestone is visualized by the rhombus at the bottom of page 1. And the ALJ, despite the staff's negative recommendation in that case, held that no import ban should be ordered owing to the anticompetitive effects of Qualcomm's anti-Intel efforts.

The issuance of the RD in late September triggered further public-interest briefing in accordance with 19 C.F.R. § 210.50(a)(4), as shown in the upper part of page 2 of the PDF.

The statute refers to government agencies, the general public, and the parties, but those submissions are on different schedules: the parties are obviously served the RD, and then have 30 days to comment on the public interest, while the clock technically begins to tick for the general public only after an official Commission notice appears in the Federal Register. In the first Qualcomm v. Apple case, the parties had their deadline on Halloween (see Friday's post for the filings and my related commentary), while government agencies and the general public have until Thursday (November 8). The deadline for government agencies and other stakeholders is another reason for which I wanted to explain the procedures here.

The parties are challenging the parts of the ALJ's ID (on the alleged violation) that are unfavorable to their respective interests, with Qualcomm requesting a review only in the event that Apple's petition succeeds (otherwise Qualcomm would rather prevail on one patent now than, hypothetically, over up to three patents later). The Commission now has to determine whether to review the ID. In cases where, unlike here, the ALJ found no violation, a decision not to review means that the ALJ's finding of no violation is affirmed, and then there's no public interest discussion anymore unless the complainant successfully appeals.

If the Commission decides to review, there's another chance for everyone to comment on the public interest (though it obviously wouldn't make sense for those who commented before to write something again unless there's something new to take into consideration).. As the ITC says on its website, "the rules continue the Commission’s past practice inviting the parties, government agencies, and members of the public to submit comments on the public interest after the Commission decides whether to review the ALJ’s decision on violation (unless the Commission affirms an ALJ's decision that there is no violation)."

Page 2 of the diagram ends with the final Commission determination, but if that one involves a finding of a violation and an exclusion order (import ban), it still isn't over. Not only can there be an appeal, but more immediately, the Commission will write a letter to the United States Trade Representative (USTR), to whom U.S. presidents typically delegate the review of ITC import bans. The Presidential review period takes 60 days, and no import ban takes effect before the President's office (in practical terms, the USTR) has had the chance to veto it. As page 3 shows, the USTR also receives briefing from the parties--and third parties will usually try to lobby the federal government as well.

In closing I'd just like to explain the colors in the diagram:

  • white background = ITC and USTR

  • red = complainant

  • green = respondent

  • yellow = complainant and respondent (as yellow results from mixing red with green)

  • blue = government agencies and general public

I hope that diagram, as well as my new smartphone patents battlemap, will prove useful to many of you as these processes unfold!

Share with other professionals via LinkedIn:

Saturday, November 3, 2018

Wall Street sees 20%-25% regulatory risk for IBM's envisioned acquisition of Red Hat

Last Sunday, IBM and Red Hat announced a merger agreement under which "Big Blue" (NYSE:IBM) would pay $34 billion, or $190 per NYSE:RHT share, to acquire the company that once started as a Linux distributor.

I may very well talk about the strategic ramifications of the proposed transaction some other time, but the focus of this post is exclusively on what the stock market appears to think of the deal.

On Monday (October 29), Bloomberg already reported on what was then a 12% spread, "among the highest for North American deals." The article quoted a portfolio manager who said he didn't want to bet on a deal that may be about a year away from closing, and IBM CEO Ginni Rometty as denying "any regulatory inhibitors," which she obviously had to say.

The time frame certainly affects demand, given that risk arbitrageurs could in the meantime use the money they would spend on RHT shares now to bet on a couple of other mergers, provided that those other deals would close more quickly and happen sequentially. But there's more to it. The spread does indicate that merger-focused investors are far from convinced that the deal will materialize.

On Friday (November 2), RHT closed at $172.24. If the deal went through, those investing now would then rake in a profit of more than 10%. Even if it took a year, a 10%+ gain would be a great deal. The only explanation for why there isn't stronger demand, at a higher price, is skepticism. Since I can't imagine anyone doubts that IBM is a serious buyer, the reason must be concern about the merger review process in the U.S. (DoJ), EU (European Commission, DG COMP), and China (MOFCOM). While China prevented Qualcomm from acquiring NXP, IBM reportedly claims it's not critical for the Red Hat deal. I haven't formed a definitive opinion on it yet, but for now I'll take IBM's word for it.

Without going into detail (yet) on the issues presented by the transaction, we can "reverse-engineer" the stock price in order to get an idea of how likely the deal is--in the eyes of sophisticated Wall Street investors--to go through or fall through. Let's start with the roughest and simplest approach, and then fine-tune it a little bit to take the time value of money into account.

The potential upside based on Friday's closing price is $17.76. Theoretically it's even greater since someone else might try to outbid IBM, but that doesn't appear to be considered too likely by anyone.

The potential downside here would not realistically be a complete loss of the investment. Red Hat is doing too well to go out of business anytime soon. The closing price over which IBM offered a premium of about 60% was $116.68, pretty much at a level with RHT's 52-week low of $115.31 (an intraday price as far as I can see).

If the deal falls through, it's a reasonable assumption that Red Hat's stock price will go back to that level, though it's obviously hard to predict what the market environment would be at that point in time somewhere in the second half of 2019. In order to keep things simple, let's not consider that investors might think they could mitigate their loss by getting out once there's a serious negative sign, such as a powerful Statement of Objections (SO) by the European Commission.

Assuming that the pre-merger-announcement closing price is where the price would fall, the potential per-share loss is $55.56 ($172.24 - $116.68).

If the likelihood of closing is estimated to be 76%, and the risk of things falling apart is (consequently) 24%, then there would be an equilibrium (76% x $17.76 is at a level with 24% x $55.56).

Let's fine-tune this by assuming a financing cost of $4.00 per share (roughly the Fed rate, assuming that you have this cost for 12 months). In that case, the potential gain (by placing the right bet on the deal going through vs. doing a far safer investment that would have a 2.5% yield) is $13.76, and the potential loss increases to $59.56. There would then be an equilibrium if the risk of the deal being blocked (or remedies being imposed with the effect of the deal falling through) was estimated at 19% (19% x $59.56 is at a level with 81% x $13.76).

Even in the aftermath of Qualcomm-NXP, that is a fairly skeptical perspective, given that mergers of this kind normally go through.

I've received a couple of independent invitations to meet portfolio managers in New York to discuss the deal, plus various calls and emails. Investors used to follow my coverage of the Google-Motorola process with great interest, and many still remember my vocal opposition to Oracle's acquisition of Sun Microsystems in late 2009 and early 2010. It was funny for me how the numerous Wall Street people I talked to always called the company "JAVA" (based on the stock ticker symbol). Java--the programming language--wasn't an issue at the time; MySQL, the open-source database, was the reason for which the European Commission conducted a Phase II review and issued an SO. It's about open source again, and this time around, Java will be part of the discussion.

As a matter of transparency, Red Hat contributed to my NoSoftwarePatents campaign in late 2004 and early 2005 (two other companies, including one far smaller than Red Hat, were much bigger supporters of the campaign), but there hasn't been any business relationship with Red Hat since. I've never had any relationship with IBM, other than having the greatest respect for the work of its patent department.

Share with other professionals via LinkedIn:

Friday, November 2, 2018

Apple stresses ramifications for 5G in public interest statement filed in Qualcomm's ITC case

On Halloween, shortly after the United States International Trade Commission (USITC, or just ITC) released a public redacted version of Administrative Law Judge (ALJ) Thomas B. Pender's initial determination (ID) regarding Qualcomm's first ITC complaint against Apple, the parties' public interest statements were due.

Let's talk about procedures first. The ID on the merits must be viewed separately from the recommended determination (RD) on remedies. On the merits, ALJ Pender sided with Qualcomm on one of three remaining patents-in-suit, but not on the other two. But even to the extent Qualcomm might ultimately prevail on the merits, the ALJ spoke out against an exclusion order in light of the anticompetitive effect this would have in light of Intel being the only Qualcomm competitor in terms of a baseband chipset maker willing to sell to actual customers as opposed to just making chips for its own devices (such as Huawei's HiSilicon).

  • Apple petitioned the Commission (the ultimate decision-making body of the U.S. trade agency) to review ALJ Pender's ID on the merits and find no infringement of a valid patent.

  • While I would have thought Qualcomm would be disappointed at prevailing on only one out of six originally asserted patents and seek to prevail on one or two more patents, Qualcomm is apparently optimizing for time: it wants to get a potential import ban sooner rather than later. Therefore, Qualcomm would take the bird in the hand instead of three birds in the bush, and filed a contingent petition: ideally Qualcomm would like the Commission to decline the ID on the merits, but in case it does, Qualcomm wants the chance to prevail on the other two remaining patents-in-suit.

    Qualcomm's strategy is actually consistent with how it treated certain patents in the San Diego litigation with Apple, where Qualcomm simply waived its rights to assert a whole bunch of patents in order to avoid a finding on the merits, particularly with respect to patent exhaustion. Before anything takes longer or before a strategically undesirable issue (exhaustion) gets adjudicated, Qualcomm would rather throw some patents away because it owns so many more.

    The Office of Unfair Import Investigations (OUII, commonly referred to as "the ITC staff") disagrees with both petitions.

  • The parties cannot formally petition the Commission to review the RD on remedies. However, the Commission requested (new) public interest statements. I published that request at the very end of my post on the ID.

    Those statements were filed on Halloween. I saw Qualcomm's statement yesterday, but Apple's was sealed. Today I was able to download a public redacted version of Apple's statement.

Let's start with Qualcomm's statement since Qualcomm is the de facto "appellant" of the RD on remedies (this post continues below the document):

18-10-31 ITC 337-TA-1065 Qualcomm Pub Int Stmt by Florian Mueller on Scribd

Qualcomm says Apple and Intel work "hand-in-glove" (which may actually be the case based on how they pursue their invalidity challenges to Qualcomm's German patents-in-suit, using the same patent law firm and making near-identical filijngs), and describes ALJ Pender's RD as "unprecedented" because the ITC usually does impose U.S. import bans on mobile phones found to infringe patents not deemed invalid, because all other ALJs who conducted fact finding on the public interest (after the Commission delegated this to ALJs, starting in 2011), and because "[t]here has not been a case since the 1980s in which the Commission has found a violation of Section 337 without issuing an exclusion order, and in each case, there were no available substitutes for the excluded products."

But all of what Qualcomm says in that regard depends on its choice of the relevant market: smartphones. The ALJ, however, determined that the public interest issue in this case is about the market for cellular baseband chipsets--more, specifically, baseband processors for the premium segment, and focusing on "merchant" suppliers (Qualcomm and Intel) as opposed to chipset makers that more or less exclusively supply their chips only to other companies of the same group (HiSilicon supplying chips to Huawei, for instance).

Qualcomm argues that focusing on the effect an import ban would have on Intel's viability as a Qualcomm competitor would make it impossible for the ITC to "enforce any patent against any iPhone with an Intel chipset." But in this context Qualcomm doesn't mention its own tactical choice to target only Intel-powered smartphones through its ITC complaints (and, effectively, its 13 patent assertions in Germany).

Apple's public interest statement highlights (in its first footnote) a fact that shows Qualcomm's targeting of Intel-powered iPhones through ITC complaints can't simply be attributed to technical considerations:

"Underscoring the discriminatory nature of Qualcomm's requested remedy, Qualcomm alleges infringement of U.S. Patent No. 8,633,936 based on graphics processing functionality in the accused iPhones that has nothing to do with the baseband chipset."

Now I've quoted Apple's statement before showing it (this post continues below the document):

18-10-31 ITC 337-TA-1065 Apple Pub Int Stmt by Florian Mueller on Scribd

Apart from footnote 1 (quoted above the document), a couple of aspects of Apple's filing are worth highlighting:

  • Apple attributes to Qualcomm's "anticompetitive practices" the fact that "in the past decade, multiple U.S. baseband chipset suppliers--including Broadcom, Marvell, and Texas Instruments--exited the [relevant] market." This focus on U.S. chipset makers (while there would also be examples in Asia) makes sense since the ITC is concerned about protecting U.S. companies from unfair competition. It's an "America first" agency (though there are many Democrats there who won't like that slogan).

  • "Except for Intel's portion of sales to Apple, Qualcomm is the sole merchant supplier to every other manufacturer supplying premium smartphones in the United States."

  • Unlike Qualcomm, Apple places a lot of emphasis on 5G. Basically, the argument is that U.S. leadership in 5G depends on Qualcomm facing at least one major U.S. competitor (Intel). Apple points to the Presidential Memorandum on Developing a Sustainable Spectrum Strategy for America's Future issued on October 25, 2018.

    Actually, Qualcomm also mentions the President: Qualcomm argues that national security considerations (related to 5G) are not part of the Commission's public interest inquiry but might later give rise to a presidential veto.

  • Just like ALJ Pender, Apple argues that Qualcomm could still obtain damages in the companion case filed in district court. While this is a simple procedural truth, it's important not to forget that if the ITC denies an import ban, it doesn't mean that anyone gets away with an infringement or that a patent holder wouldn't be compensated.

Share with other professionals via LinkedIn:

Thursday, November 1, 2018

FTC chairman recused from Qualcomm antitrust litigation: The Capitol Forum

The Capitol Forum (with whom I'm connected on social media) has just broken the following news:

"Qualcomm: Simons Recused from FTC v. Qualcomm Litigation; Democrats at FTC Have De Facto Veto Power Over Consent Agreement"

The subscription service cites "sources familiar with the matter." They contacted spokespeople for two FTC commissioners and Qualcomm, but they declined to comment or elected not to respond.

One of the sources said the remedy resulting from a settlement could involve arbitration. It's unclear what exactly would be arbitrated. The question of whether Qualcomm must extend a license to rival chipset makers such as Intel is binary and does not lend itself to arbitration, and I just recently, in connection with Huawei v. Samsung, reiterated my longstanding position on coerced FRAND arbitration.

I don't know what exactly caused Chairman Simons's recusal. I do, however, have concerns about the ties of certain key FTC officials (not commissioners, but nevertheless highly influential) with Qualcomm.

While I can imagine why the Capitol Forum's headline puts the chairman's recusal into the context of a potentially partisan vote, I absolutely positively don't view this here as a left-right, conservative-liberal, Republican-Democrat issue. Case in point, no senator was a more positive force in the previous fight against SEP abuse (when Motorola Mobility and Samsung were the culprits) than Sen. Mike Lee (R-Utah). I would recommend to any Republican concerned about the Qualcomm case being a "Democrat" affair to read some of the letters and listen to some of the speeches Senator Lee gave at the time.

No IP/antitrust blogger I know supported Donald Trump's campaign as early as I did in a January 2016 blog post. On social media, I've also declared myself a supporter of such people as "The People's Sheriff" David Clarke and a big Rush Limbaugh fan. I expressed my disappointment over Judge Robart's factually and legally erroneous positions on the travel ban, and supported the #ConfirmGorsuch and #ConfirmKavanaugh campaigns. Believe me, if I thought there was anything leftist or heavy-handedly statist (a term Rush also uses from time to time) about the FTC case against Qualcomm, I'd have said so. But there simply isn't.

An important matter for U.S. innovation and consumers shouldn't be stigmatized because the FTC filed the lawsuit in the waning days of the Obama Administration. It's a just cause regardless. I'd be a hundred million times more concerned about some other things Obama and his people did shortly before leaving office, such as sending $221 million to the Palestinian Authority, which is known to make payments to the surviving family members of "martyrs," or more accurately, suicide terrorists.

The United States is the cradle of antitrust law, and the Qualcomm case, which involves three major U.S. companies (Qualcomm, Apple, Intel), is a key case and an opportunity for the U.S. to assert and maintain its thought leadership.

With a view to future elections, including President Trump's reelection, I also think it would actually be better for Republicans--of course to the extent that the law and the facts so merit--to pursue the Qualcomm matter since a successful FTC trial would pave the way for a payout to 250 million Americans. In the alternative, if the FTC settled but the class action succeeded regardless, it would look as if Republicans didn't want people to get some of their money back that they overpaid on past smartphone purchases.

In 2009, Rush Limbaugh famously wanted President Obama to fail, a statement that was widely misunderstood. Now I just find it ridiculous how the former president claims credit for the fruits of President Trump's #MAGA economic policies. Assuming the FTC v. Qualcomm trial goes forward and succeeds, I don't think Democrats would be able to get any mileage out of it regardless of when the complaint was filed.

Let us hope that any vote held by the FTC will be driven by fact-specific considerations and principles that are important not to protect companies from the market, but to protect consumers from egregious, abusive conduct.

Share with other professionals via LinkedIn: